A graph neural network-enhanced knowledge graph framework for intelligent analysis of policing cases.

In this paper, we model a knowledge graph based on graph neural networks, conduct an in-depth study on building knowledge graph embeddings for policing cases, and design a graph neural network-enhanced knowledge graph framework. In detail, we use the label propagation algorithm (LPA) to assist the convolutional graph network (GCN) in training the edge weights of the knowledge graph to construct a policing case prediction method. This improves the traditional convolutional neural network from a single-channel network to a multichannel network to accommodate the multiple feature factors of policing cases. In addition, this expands the perceptual field of the convolutional neural network to improve prediction accuracy. The experimental results show that the multichannel convolutional neural network's prediction accuracy can reach 87.7%. To ensure the efficiency of the security case analysis network, an efficient pairwise feature extraction base module is added to enhance the backbone network, which reduces the number of parameters of the whole network and decreases the complexity of operations. We experimentally demonstrate that this method achieves a better balance of efficiency and performance by obtaining approximate results with 53.5% fewer floating-point operations and 70.2% fewer number parameters than its contemporary work.

An unprecedented spike of the electroluminescence turn-on transience from guest-doped OLEDs with strong electron-donating abilities of host carbazole groups.

An unreported unprecedented spike of ∼µs line-width, followed by an overshoot, was discovered at the rising edge of transient electroluminescence (TEL) from guest-doped organic light-emitting diodes with strong electron-donating abilities from the host carbazole groups. By changing the device structures and TEL measurement parameters, a series of experimental results demonstrate that this TEL spike is not related to exciton interactions such as singlet-triplet and triplet-triplet annihilations but originated from the radiative recombination of pre-stored electrons with injected holes. Surprisingly, these pre-stored guest electrons do not come from the energy-level traps in the host-guest systems; instead, the guest molecules receive the electrons transferred from the host carbazole groups due to their strong electron-donating abilities. Moreover, the observed spikes show rich and extraordinary temperature dependences. Based on the detailed understanding of the spike formation mechanism, we have proposed the requirements for the occurrence of spike and realized the artificial adjustments of the spike intensity. For instance, the instantaneous luminescent intensity of this spike can reach over 80 times the magnitude of the TEL plateau. Accordingly, this work deepens the physical understanding of this novel spike in TEL and paves the way for fabricating an electro-optic sensor to detect instantaneous weak current signals.

New Insights into the Hourly Manure Coverage Proportion on the Manure Belt in a Typical Layer House for Accurate Ammonia Emission Modeling.

The main advantage of having livestock, for example, the laying hens, in a controlled environment is that the optimum growth conditions can be achieved with accuracy. The indoor air temperature, humidity, gases concentration, etc., would significantly affect the animal performance, thus should be maintained within an acceptable range. In order to achieve the goals of precision poultry farming, various models have been developed by researchers all over the world to estimate the hourly indoor environmental parameters so as to provide decision suggestions. However, a key parameter of hourly manure area in the poultry house was missing in the literature to predict the ammonia emission using the recently developed mechanistic model. Therefore, in order to fill the gap of the understanding of hourly manure coverage proportion and area on the manure belt, experimental measurements were performed in the present study using laying hens from 10 weeks age to 30 weeks age. For each test, six polypropylene (pp) plates were applied to collect the manure dropped by the birds every hour, and photographs of the plates were taken at the same time using a pre-fixed camera. Binary images were then produced based on the color pictures to determine the object coverage proportion. It was demonstrated that for laying hens of stocking density around 14 birds/m2, the manure coverage proportion at the 24th hour after the most recent manure removal was about 60%, while the value was approximately 82% at the 48th hour. Meanwhile, for laying hens at different ages, the hourly increment of manure coverage proportion showed a similar pattern with four distinct stages within 48 h. The statistical analyses demonstrated no significant correlation between the hourly increment of manure weight and the hourly increment of manure coverage proportion. Finally, prediction models for estimating the hourly manure coverage proportion on the manure belt in typical laying hen houses were provided.

Achievement of High-Level Reverse Intersystem Crossing in Rubrene-Doped Organic Light-Emitting Diodes.

Using the fingerprint magneto-electroluminescence trace, we observe a fascinating high-level reverse intersystem crossing (HL-RISC) in rubrene-doped organic light-emitting diodes (OLEDs). This HL-RISC is achieved from high-lying triplet states (T2,rub) transferred from host triplet states by the Dexter energy transfer to the lowest singlet states (S1,rub) in rubrene. Although HL-RISC decreases with bias current, it increases with lowering temperature. This is contrary to the temperature-dependent RISC from conventional thermally activated delayed fluorescence, because HL-RISC is an exothermic process instead. Moreover, owing to the competition of exciton energy transfer with direct charge trap, HL-RISC changes nonmonotonically with the dopant concentration and increases luminous efficiency to a maximum at 10% of rubrene, which is about ten times greater than that from the pure-rubrene device. Additionally, the HL-RISC process is not observed in bare rubrene-doped films because of the absence of T2,rub. Our findings pave the way for designing highly efficient orange fluorescent OLEDs.

Spin-pair state-induced exceptional magnetic field responses from a thermally activated delayed fluorescence-assisted fluorescent material doping system.

The thermally activated delayed fluorescence (TADF) material 2,3,5,6-tetrakis(3,6-diphenylcarbazol-9-yl)-1,4-dicyanobenzene (4CzTPN-Ph) and the conventional fluorescent dopant 4-(dicyanomethylene)-2-tert-butyl-6-(1,1,7,7-tetramethyljulolidin-4-yl-vinyl)-4H-pyran (DCJTB) were used to co-dope the host material 4,4'-bis(carbazol-9-yl)biphenyl (CBP) for the fabrication of TADF-assisted fluorescent organic light-emitting diodes (OLEDs). Some exceptional magnetic field effect (MFE) curves with abundant structures and four tunable components within a low magnetic field range (≤50 mT) were obtained, in sharp contrast to the maximum of two components observed in typical OLEDs. These MFE components were easily tuned by the injection current, dopant concentration, working temperature, and dopant energy gap, leading to a wide variety of MFE curve line shapes. The experimental results are attributed to the spin-pair state inter-conversions occurring in the device, including intersystem crossing (ISC) of CBP polaron pairs, ISC of 4CzTPN-Ph polaron pairs, reverse ISC (RISC) of 4CzTPN-Ph excitons, RISC of DCJTB polaron pairs, DCJTB triplet fusion, and DCJTB triplet-charge annihilation. Moreover, the exciton energy transfer processes among the host material and the guest dopants had a pronounced impact on the formation of these four components. This work gives a deeper understanding of the microscopic mechanisms of TADF-based co-doped systems for the further development of organic magnetic field effects in the extensive field of OLEDs.

Using magneto-electroluminescence as a fingerprint to identify the spin polarization and spin-orbit coupling of magnetic nanoparticle doped polymer light emitting diodes.

The spin polarization and spin-orbit coupling (SOC) in polymer light emitting diodes (PLEDs) with the active layer doped with Fe3O4 nanoparticles (NPs) were identified through magneto-electroluminescence (MEL). By comparing the MEL characteristics such as linewidth and magnitude between PLEDs with and without Fe3O4 dopant, we confirmed the existence of spin polarization, but ruled out the existence of SOC. Although the spin polarization is positive to electroluminescence, the brightness-current characteristics suggested that the current efficiency of the doped PLED does not improve. We attributed it to the current leakage caused by the Fe3O4 NPs in the active layer. This work is beneficial for us to further understand the effect of magnetic nanoparticle doping on the dynamic behavior of excitons and polaron pairs in organic semiconductor devices.